antenna-theory.com

[gert3d]

I am testing out a monopole antenna for a simple 433MHz transmitter/receiver combination with an Arduino.

I found a page on this website on monopoles:
http://www.antenna-theory.com/antennas/monopole.php

I assume, although it is not mentioned in the text or detailed in figures, that the a mounted monoploe antenna is isolated from the ground plane?
What is the benefit of having a ground plane for a vertical monopole, since it is mentioned at the end:

"as the ground plane approaches infinite size, the radiation pattern approaches a maximum in the x-y plane"

which is the same characteristic as without a ground plane?

In my measurements I find optimum transmission between lengthts of a free (1mm copper) monopole of 19 - 23 cm. Different posts say it should be 17.5 cm. (supposedly 1/4 of wavelength). What influences the transmission efficiencies other than length of the vertical free standing monopole?

Thanks for answers ...

[Schubert]

Ok. For an antenna, we have two sides of every antenna. We can think of the source as a voltage source with a positive and negative (ground) side. For the dipole antenna, the voltage source has one arm of the dipole connected to the positive side, and the opposite arm of the dipole connected to the negative side of the voltage source. The dipole arms are isolated from each other outside of thsi connection through the source.

For the monopole, we have one arm above the ground plane as in the dipole case, which is connected to the positive side of the voltage source. The ground plane makes up the opposite arm of the antenna, which is connected to the negative side of the voltage source.

What is the benefit of this? Well if you are in the middle of space and need to make an antenna, then a dipole will be just fine. But let's say you want to place an antenna on top of an airplane, for instance. Then you have a large ground plane avaialble (the chassis of the metallic surface of the airplane). Hence you can use this as one arm of your antenna. Then you only need a quarter-wavelength dipole arm above the ground plane and you have a nice antenna. Hence, monopoles are useful for attaching antennas to structures, where you can use the existing structure as part of the antenna.

Then you write:
"In my measurements I find optimum transmission between lengthts of a free (1mm copper) monopole of 19 - 23 cm. Different posts say it should be 17.5 cm. (supposedly 1/4 of wavelength). What influences the transmission efficiencies other than length of the vertical free standing monopole? "

It depends on what frequency you are operating at. Antenna lengths are designed for given frequencies. Hence a 19cm antenna will be fine for one frequency but not another.

[gert3d]

Thanks, that is clear. Gives me more ammunition for further tests ...

I did further measurements with 433MHz transmission and reception using simple monopoles. By lowering the voltage of the transmitter I was able to do measurements with more discriminative values and indeed observed highest efficiency between 17 - 19 cm of monopole antenna.

Under these circumstances the monopole of the receiver had a much wider optimum: between 14 - 32 cm.

I have been using this:
http://www.evola.fr/product_info.php/en/link-kit-433mhz-arduino-compatible-p-162
and tested transmission efficiency at lowest possible voltage (1.6V) over 30 meters, including walls.

[E Kafeman]

Monopole with highest efficiency have antenna length 0.25 wavelength in vacuum. In normal air is transmission speed a bit slower and conductors are not ideal, so 0.24 wavelength is a commonly accepted value for an optimal monopole length. This assumes a perfect ground, infinite size or at least a circle diameter of 350 mm in this case.
Any other length will reduce antenna efficiency in horizontal plane.
However, to get max performance, with lowest possible losses, must antenna feed (PCB trace/coax cable) be of correct impedance. Also must TX/RX circuit be matched for this impedance. An impedance mismatch causes high VSWR =>RF energy is reflected back and causing heat in TX circruit and less energy is delivered to RX circuit.
Assume that RX circuit is designed for 200 Ohm. It would be a mismatch against an ideal dipole, 73 +j43 Ohm. If dipole length then is shortened will total antenna impedance increase => somewhat better match but in total a less effective system. It is always better to adjust RF circuit to avoid losses and make it impedance matched against an antenna with correct size. A correct RX impedance would in this case be 73-j43 Ohm.
Both antenna and active radio circuit are commonly adjusted for 50 +j0 Ohm. That impedance adjustment can be done with inductors and capacitors, coaxial or PCB trace stubs or a lot of other methods. Much of that information can be found on this site.

If you find that 0.24 wavelength not results in best coverage for an monopole antenna connected to a reasonable ground, is the main problem TX/RX feed impedance to the antenna, not the antenna in it self even if a misadjustment of radio impedance partly can be improved by a misadjusted antenna. In total is it however always a less effective system compared to a radio and antenna that is correct tuned.

Most effective radio system do not need to be the same as the system with longest coverage!! Two examples:
1. If it is very strong RF signals say at 420 MHz from other transmitters, can it over-saturate your receiver at 433 MHz if it is feed with too high levels and make it deaf for the weak signals that you actually want to receive. RX coverage at 433 MHz can then increase by reduce antenna efficiency as it then also reduces unwanted signals, making total system less deaf.
2. An antenna can be mistuned if placed close to a wall, reducing total efficiency, but the wall can also cause antenna directivity that increases coverage in one direction, compensating for more then the mistune cost, even if it in total is a loss.

[gert3d]

@ Kafeman : Thanks! Explains a lot, more than I can optimise I am afraid, but gives direction for my thoughts.

[E Kafeman]

You can optimize further with very simple tools. With aid of a diode detector, a 50 Ohm resistor and a multimeter can impedance be corrected for both radio and antenna. Finding optimal impedance correction values can be done with variable capacitors and inductors, and when found , replaced with fixed values.
It requires a bit trial and error to find optimal values but it is possible.

A diode detector is built in 5 minutes and is very useful for both checking radio circuits as well as antenna function. In combination with the resistor can antenna be tuned like this:

Build a stable antenna with correct length. A stiff iron wire placed on main PCB is ok, a flexible wire will not work. Place a transmitter at a few meters distance.
Disconnect radio circuit from antenna.
Connect the 50 Ohm resistor instead of the radio circuit. Measure resulting rectified voltage at the 50 Ohm resistor, use 1 kOhm SMD resistors as rf decoupling between detector and 50 Ohm load, for avoiding losses/mistuning. Keep wires and other stuff away from antenna and keep all connections short and close to ground, avoiding that measurement equipment introduces RF-related errors.
Make place for a L-network between antenna and 50 Ohm resistor with shortest possible PCB trace length.
Place a adjustable cap 1-10 pF in serial => If no improvement it should be an inductor. Shortcut until further and move the cap so it now connect from antenna trace to ground. Tune and check if any improvement. If not, similar test must be done with inductors. I do that test with fixed inductors and a lot of trial and error as it must be either a cap or a ind.
If a cap, check that he variable cap have enough range, or else add serial and parallel capacitors until tuning range is within limits and can be estimated based on capacitor tuning angle.

Now go back to the serial component and find optimal value once again. That adjustment will cause that component connected to ground need to be adjusted, again. Repeat and repeat until no further improvements can be found within E12 values.

Note component values and remove them! and replace antenna with the 50 Ohm measurement tool and measure directly at the local transmitter.

Do the trial and error procedure again, to find optimal values for this L network.

Note the values.
Now do you have enough information to be able to calculate both TX circuit antenna complex impedance at the point where L network is located, as you have a 50 Ohm match there and a discrete matching net in both directions, with known values.
Either just connect these both nets in serial or recalculate components to one single network that fulfills matching in both directions. It is no need to have 50 Ohm impedance in any certain point of the transmission trace and it saves 1 or two components that else cost some losses as it not is ideal components.

Now do a double check by adjusting each matching component one step up or down, but now with complete populated circuit, battery and plastic enclosure or what its final environment will be as that also is a part of total antenna impedance and now need to be readjusted. This time with the diode detector placed at distance and connected to an other antenna or a short wire. This antenna do not need to be tuned at all.
Distance of a one or two meters is enough to avoid that measurement wires causes local mistuning. Reflective environment can make this tuning unstable. Typical, if someone is moving around in same room or it is hard to find a stable correlation between antenna distance and measured signal level. In worst case must this final tuning be done on wooden table, placed on a football plane or a similar reflection free space.
Anechoic chamber is also fine as measurement range but it is not an alternative if the tuning should be done in a low budget.

This kind of tuning takes in total few hours. It goes much faster measuring with a Vector Network Analyzer but it costs a bit more then a diode detector. $100k or more is common. If same thing can be done with tools for $1 is it good saving pro hour.

I am sure that you are able to do this optimizing and measuring around with a diode detector gives very valuable RF and antenna insights. I would not be surprised if max coverage distance can be doubled.

[gert3d]

Thanks, that is worth a try!

I'll just have to do it to understand the principle completely. A few questions before I start:

1. The diode detector. What is that exactly? I do not see it used in the testing procedure.
2. The 50 ohm is placed between antenna and ground instead of receiver?
3. What should the multimeter measure for optimum adjustments? zero Volts?
4. will it work testing low power 433MHz transmitters as well? The transmitter only takes 40mA @ 5V.

Thanks

[E Kafeman]

1. It is a simple form of RF rectifier. Resulting output can be measured as a voltage or current.http://www.astarmathsandphysics.com/a_level_physics_notes/electronics/a_level_physics_notes_electronics_am_modulation.html
As we also add serial resistors does it not add any RF load more then what the 50 Ohm resistor represent.
2. Yes, the 50 Ohm resistor is assumed a ideal 50 Ohm load, which is sensed by the antenna. We can not use the receiver as load for obtaining correct tuning yet, as we not know the RX input impedance.
3. Max voltage or current, depending on what range that is most sensitive. It is often preferred to use an mechanical analog instrument as a needle is easier to read if it moves in any direction.
4. Tuning above was for antenna and Tx. As Rx often is a part of same chip as Tx is it not easy to optimize separately for Rx and Tx. If RSSI reading is available can RX impedance matching be done in similar way.
If output TX level is low can it be needed to add active amplifying.
RF active probe:http://www.next.gr/mobile/digital/logic-circuits/active-rf-detector-probe-l14326.html